Professional Documents
Culture Documents
Invariance:
•“External parameters”
•Pose
•Illumination
•“Internal parameters”
•Person identity
•Facial expression
Task
Given an input image,
determine if there are
objects of a given class (e.g.
faces, people, cars..)
in the image and where they
are located.
Off-line
training
Classifier
Feature Extraction
Training Set
Train Classifier
False Positive
Sensitivity
Classify
19x19
Histogram Masking
Equalization
Gray 283 [0, 1]
x-y-Sobel
Histogram
Filtering
Equalization
Gradient 17x17 [0, 1]
Real Synthetic
2900 faces + 2900 mirrored faces (T. Vetter, Univ. of Freiburg)
Illumination
Identity
Rotation
3D Morphing: + =
Add to
training set
Bootstrapping
Rotation out of
image plane
• Component-based classification
Rotation in the • Train on rotated faces
image plane
Single
template
Component
templates
2nd Level:
Geometrical
relation
between
components classify
maximum response
of each component
classifier + x, y
location
Components:
• discriminatory
• robust against changes in
pose and illumination
Synthetic faces:
• 7 different 3-D head models
• 2,500 faces
Rotation: -30o to + 30o
• 3-D correspondences
for automatic location of
components
x *2
Bound on error
M 2
E < c(R / M)
R
M
x*1
Feature Space
Cross V
alidation might be better
0.9
0.8
0.7
0.6
Correct
0.5
14 learned components
0.4
whole face
0.3
0.2
0.1
0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
FP / inspected window
Heisele, B., T. Serre, M. Pontil, T. Vetter and T. Poggio. Categorization by Learning and Combining Object Parts.
In: Advances in Neural Information Processing Systems (NIPS'01), Vancouver, Canada
Stan Bileschi
Fall 2002 Pattern Recognition for Vision
Training on Faces
Red:
Trained
only with
faces.
Blue:
Trained
on faces
and non-
faces.
A B
Task:
Given an image of an
A B
object of a particular class
(e.g. face) identify which
exemplar it is.
C D
C D
1. Multi-class problem
Illumination Rotation
Identification Result
Training Data
A
Face Image
A or B or C or D A B,C,D
A or B C or D B A,C,D
C A,B,D
D A,B,C
A B C D
Training: L
Training: L (L-1) / 2
Classification : L
Classification : L-1
Max Operation
Identification result
Identification result
Max Operation
Identification Results
Feature 1 != Feature 2
Feature 1 = Feature 2
Jennifer Huang
Heisele, B., P. Ho and T. Poggio. Face Recognition with Support Vector Machines: Global Versus Component-based
Approach, International Conference on Computer Vision (ICCV'01), Vancouver, Canada, Vol. 2, 688-694, 2001.
Jennifer Huang
Jennifer Huang
Problems Encountered:
Detection
Inaccurate Component detection
Recognition
Accuracy of 3D models
Choice of Illumination and Pose
Jennifer Huang